Professional Documents
Culture Documents
net/publication/350721483
CITATIONS READS
6 1,611
3 authors, including:
All content following this page was uploaded by Vikranth Kadya on 08 April 2021.
© With Author.
Indian population. In India, agriculture is one of the broadest sectors contributing to the Indian’s
economy, culture and social domicile. The current methodologies and practices used in the system of
agriculture trails the sector in all dimensions. By 2050, the entire population of the world reaches nearly
10 billion, so we should ask ourselves some research questions about food safety and agricultural
production [1]. To meet these requirements, agricultural production must increase by 60 percent. It is
not sufficient and consistent to achieve that the traditional method of agriculture. In addition, now some
experts are starting to incorporate innovations to combine with the practice of agriculture with the idea
of low costs and better performance. IoT and Data Analytics play a significant role in agriculture as a
result that it leads innovations such as Wireless Sensor Networks (WSNs). The development of
technology and global networking helps agriculture solve problems systematically and it gives the label
of IT agriculture or smart farming. Sensors also play a crucial role here, gathering various crop
information such as information on moisture, insect information, whether information, data on water
level, soil information and field information. To collect all this information, various types of sensors
such as moisture sensors, electromagnetic sensors, and optical sensors need to be used. Such sensor-
enabled devices produce a big sized data that can be placed in a different storage system or cloud
database. Agricultural big data solutions include technologies such as Hadoop Distributed File System
(HDFS), Map Reduce, STORM and so on. So if we implement all of these technologies in agriculture,
it will reduce crop failure and satisfy farmers' primary concern to generate a profit. Adopting these
technologies enhances the farming cycle by optimizing the efficiency and reducing environmental
effects. This also helps farmers in how much fertilizer to use, and water requirement, and pesticide and
insecticide information [2]. Wireless Sensor Network (WSN) is a state-of-the-art technology comprised
of sensor comprehension, mechanization manage, digital network transmission, information storage,
and information dispensation to provide effective solutions for a target range of farm applications. The
efficiency, productivity, and profitability of many agricultural production systems by using the Wireless
Sensor Networks (WSNs) would increase precision agriculture. With the aid of the wireless network,
IoT helps farmers with live updates by informing people about the crops. We can connect different
sensors such as sensors for temperature, sensors for moisture, sensors for water level, and sensors for
flood gauge. With the aid of these different sensors, this can provide the farmer with conscious
notifications about what is continuing in the field [3]. The development of wireless communication
technology leads to improvements in relatively low cost, less power consumption and small-scale
multifunctional sensors that allow short-range communication [4]. IoT's architecture consisted of 3
strata. The front end layer gathers information about the environment generated by various agricultural
activities. The middle layer acts as a layer of the gateway that connects the front end layer to the Internet.
The back end layer stores all the front end data collected [5]. Collectively, the sensor nodes are useful
to transmit the précised data to the sink, called a bottom station to screen and extent a conclusion about
an atmosphere. The base station does the interpretation of the discoveries on the data collected [6]. The
word big data refer to the large number of data sets available. Sensors, phones, and other devices are all
producing big data. The primary goal of acquiring such large data source is to use efficient processing
to search for meaningful patterns. The big data collected using different sensors that undergo several
phases of arrangement, dispensation, and organization into significant information that can later be
analyzed for proper decision-making. Sensors are a critical part of Big Data [7]. It's a procedure of
inspecting uncovered arrangements, discovery an unknown correlation, and result useful evidence
accepted for analyzing decision-making. The IoT and data analytics focused on fields such as
agricultural production forecasting, manufacturing forecasting, machine learning algorithms, the
forecast of failure, predictive preservation, supply chain management, recurrent pattern mining. The
sensor tracks soil nutrients, humidity levels, weather tracking and pest detection in smart agriculture.
With these principles, we can farm sustainable without water wasting and reduce fertilizer use. The
farmers are insisted on using the mobile application in which they obtain the results of the study and
can decide to make productive agriculture on the basis of the result [8]. This paper focused primarily
on the proposed smart farming architecture, using wireless sensor networks and big data analytics. In
addition, they also reviewed some technologies used to process big data and put into practice techniques
for data analytics.
2. RESEARCH OBJECTIVES :
Our research area is on data analytics by using Wireless Sensor Network (WSN), IoT and existing
dataset to perform a smart agricultural system. For that intention, we created target-based research
questions.
How will the agricultural sector get benefited from using emerging technologies such as IoT
and data analytics?
How different IoT and data analytics solutions prove themselves to be best?
How to develop the most reliable, efficient, flexible, and cost-effective systems that meets
current agricultural needs?
What implications these findings have?
What kinds of data are needed to achieve a smart agricultural system?
How to design proposed smart-agriculture architecture?
How can IoT and data analytics be applied inside a smart agricultural system?
To know the research gap and research agendas to contribute further research.
3. METHODOLOGY USED :
The proposed study uses the method of data and information collecting to study the literature on said
topic. All the data and details that were collected are accessible online using secondary data, research
papers, books, journals, and magazines. In addition, here first all the research materials were collected
and then the analysis process was done and the common issues were identified in the whole paper, then
our own theory was established. The tool used here is the qualitative approach where attitude,
perception, and behavior are correlated with the study. The result obtained is also in non-quantitative
form. So, the focus of the work is on our insight into science and depth. Afterward, we built our own
architecture based on the knowledge learned after the review process from the reviewed paper.
4. RELATED WORK ON SMART AGRICULTURE :
Sawaitul et al. (2012) [9] identify an appropriate solution for climate predicting by introducing a new
wireless equipment, which includes a back propagation algorithm. This algorithm is applied to weather
forecasting dataset and observed the variation on other factors with respect to the change on one factor.
Neural networks are proficient of molding a weather prediction system. The neural network signal
dispensation method for weather predicting is capable of generating good yield results and can be
reflected as substitute to old climatological approaches. Jagyasi et al. (2013) [10] presents different ICT
and mobile sensing technologies that enhance the efficiency of agriculture and enable us to connect
farmers with concerned entities in the ecosystem. mKRISHITM is a technology outline for mobile
based agro-advisory facilities, which connect the countryside community with agricultural specialists
through cell phones and sensor networks. In agriculture, a microphone as a mobile sensor can be used
for sensing pest existence in the field. These pests are identified by examining the sound they make, by
using signal control, guesstimate and recognition algorithms. This supports farmers take required
measures to diminish harms due to pest attacks. Li et al. (2013) [11] address the problem by developing
a platform for data analysis of machinery used in agriculture. This research is focused on machinery
used in agriculture, such as tractors, cultivators, harvester and dryers. In the beginning, this application
collects data regarding machinery conditions, than using big data platform it analyses machinery
conditions using various indicators, then sends warning messages by prompting maintenance needs.
The data inspection inter system in this study is parallel to the data process platform, is the system
application that executes the data analytical techniques and display the results as analytical
consequences. In this study, the MapReduce data analysis engine established on Hadoop technology
and the Spark engine well-matched with Hadoop technology are generally used. Kang et al. (2013) [12]
explains weather data analytics which remotely collects the weather information and stores it in a cloud
database accessible from any location. The proposed new method of efficiently collecting data led to
new wireless sensor network applications. The concept is to use real-time sensors installed in the field
of a subject so that data can be collected in a real environment and made available for future machine
learning.
Mahmood et al. (2013) [13] elaborates on the characteristics of different BDA tools. The different
characteristics such as cloud integration, scalability, query based, parallel processing features, handling
arranged and dispersed data, freeware and Hadoop integration feature of multiple big data apparatus
such as Hadoop and MapReduce, IBM big data analytics, Kognito WX2, SAP big data analytics,
Microsoft Big data-Azure, Oracle big data analytics, Tera data big data analytics, SAS big data analytics
were discussed. Yang et al. (2013) [14] provides IOT enabled smart farming solution using data
compression and cleaning methods as well as decision-knowledge procedures. In the proposed system
the moisture data, temperature data are acquired from sensors ZigBee network. Kalman filtering reduces
noise in sensed data and performs compression. The IoT postern and cube connected on RaspberryPi
transfer the data. IoT service platform Mobius makes virtual data storage on sensed data. The decision
tree algorithm performs a forecast about crop development by means of decision learning. Here, the
decision is conversed to the user by mobile or web browser.
Jeong et al. (2014) [15] identify the solicitations and procedures of data mining in farming. It addresses
the problem of price prediction for agricultural crops. The categorization and grouping methods are two
forms of data mining techniques. For this, it can able to find appropriate approaches of data mining like
K-Means, K-Nearest Neighbor and artificial neural networks. Neural Networks and Support Vector
Machines are two categorization methods that are used commonly to categorize unidentified sample.
Here, clustering methods enables to raven a set of unknown samples into clusters. K-Means technique
is well known and frequently used clustering techniques. Data mining categorization approach can be
used to improve a pioneering form to forecast the market value of the particular goods. Medar, R. A.,
& Rajpurohit (2014) [16] explains different data mining approaches in agriculture to develop the
forecast models. The K-Means, K-Nearest Neighbor (KNN), Artificial Neural Networks (ANN) and
Support Vector Machines (SVM) are data mining approaches preliminarily used in the forecast model.
The regression method is the conjoint method for forecasting yield through a bulky area. Generally,
Neural Networks and Support Vector Machines, these two categorization approaches that studied about
categorization of unknown samples. Another classification technique, K- Nearest Neighbor, does not
have any knowledge part because it practices the training set every time a grouping must be performed.
If there is no earlier information about the data to classify than the clustering methods will be used. K-
Means algorithm is most frequently used clustering algorithm.
Parthasarathy (2015) [17] proposed the application that forecasts the environmental changes using a big
data approach which improves productivity. It presents an improved forecasting technique works
automatically based on the Hadoop framework for analysis and forecast. The MapReduce in practice of
Hadoop to monitor and analyze the big data composed from various foundations like weather
forecasting, sensor data, market tendencies and social media data. Application has 5 modules for data
acquisition, data storage, query analysis and presentation. Application projects the climatological data
storage as well as processing policy grounded on Hadoop framework with the help of linked logistic
reversion procedure for forecast. Sankaralingam et al. (2015) [18] introduced a crop simulation model
based on site specific daily weather data, soil and crop data. It is used to appraise crop vintages by
variations that happen in crop physiognomies. With the aid of FieldScout and GreenIndex farmers able
to find depiction of their crop, and the application calculates nitrogen solicitation rate commendations
constructed on a colour index, this offers planters with a low-cost technique for dealing in-season
fruitfulness which can progress yields, inferior nitrogen expenses, and upsurge earnings. Yadav et al.
(2015) [19] presents the big data that bump into small-scale sensors in precision agriculture. It gives
solutions for the big data e-agriculture service includes the prevalent current technologies such as
HDFS, Map Reduce, Hadoop, STORM, etc. Also, they analysed that the Big Data in agriculture refers
to the Electronic Farm Records (EFR) which includes data on soil, temperature, precipitation, electrical,
moisture, air, nutrient and PH levels. Several sensors can be used in real-time or in permutation with a
Global Positioning System (GPS) to create field maps of different soil properties. He concludes that the
sensing of soil and crops has turned the agricultural system into much more productive, less costly, and
better quality than ever. The spread of intelligent sensors with efficient analyses of big data will lead us
a pace on the way to freeing farmers from the restraints of unpredictable weather conditions. Forkan et
al. (2015) [20] proposed the IoT system used to senses data related to agriculture and is stored in the
remote database. The data wizard is analysed using cloud-based big data processing. Here specifications
for fertilizers, crop research, demand specifications for the crop. The prediction is then conducted using
a data mining approach in which knowledge enters the farmer via a cell phone device. Using this
predicted knowledge, our goal is to upsurge crop manufacture and governor the agricultural costs of the
goods. Li et al. (2015) [21] give the latest prominent context-conscious middleware solutions during
the period from 2009 to 2015. The IoT-based context explains awareness framework and its middleware
architecture and different context-consciousness levels. Channel et al. (2015) [22] explain a
multidisciplinary model for intelligent farming using IoT, Sensor, web Computing, Mobile Computing
and Big Data Analysis has been suggested. Farmers, Agro-Marketing companies, and Agro-Dealers
must be registered with the mobile application in the Agro Cloud network. Agro Cloud disk is used to
hold farmers' data, periodic farmland soil resources, agribusiness and agri-marketing companies, agri-
e-governance arrangements and present ecological conditions. Properties of soil and atmosphere are
sensed and occasionally transmitted to Agro Cloud via IoT (Beagle Black Bone). Agriculture cloud data
is conducted for the need of fertilizers, yield analysis, production summary, and present store and
market demands. The proposed model is helpful for cumulative agro-product progress and cost control.
Sravanthi et al. (2015) [23] suggest biotechnology companies test crops and perform simulations to
assess how different plants react to altered circumstances. They provide a thorough overview of Big
Data’s position in various areas of agriculture. Vicissitudes in the qualities of the different data it pleats,
including temperature, water echelons, soil configuration, growth, production, and gene sequencing of
each plant in the testbed. These imitations allow it to determine the ideal situations for specific types of
genes.
Ramesh, D., & Vardhan, B. V. (2015) [24] offerings a transitory analysis of crop forecast by means of
the multiple linear regression (MLR) method and Density-based clustering method for the particular
province. The foremost intention of this paper is to establish user approachable design for farmers that
provide the analysis of rice production grounded on existing data. Primarily, the numerical model
multiple line regression technique is pragmatic to present data. The consequences so gained were data
mining technique specifically the density-based clustering technique. This yield forecast model is
reachable with the exercise of the multiple linear regression (MLR) technique where the forecast is the
production and it has seven prognosticators namely year, precipitation, expanse of spreading, crop and
nourishments (Nitrogen, Phosphorous and Potassium). Area-explicit crop yield analysis is achieved by
realizing both the Multiple Linear Regression technique and Density-based clustering technique.
Ravisankar et al (2016) [25] reviewed B&M data (big data and metadata) supervision related research
in contemporary biosphere situations. Also presents design for U-Agriculture Mobile Services grounded
on Sensor-Cloud Setup. It initiates that both big data and metadata play vital protagonists in omnipresent
agriculture in handling the progressively multifaceted amount of data that are being connected with
applications or services or systems. The capability of hardware sellers that afford outfits, facilities and
podiums for dealing big. Also Microsoft is proposing mobile amenities using azure app services. Azure
App Service is a novel and exclusive cloud service that empowers developers to generate web and
mobile apps for any dais and any expedient. Stubb, M. (2016) [26] investigates how data has been
generated and collected by the farmer and other sources using current and innovative technology present
today. Normally, data gathering comprises somatic technology, such as sensors, imagery, drones, radar,
and other technologies all functioning together to deliver comprehensive evidence about earth content,
tidies and vermin, daylight and gloom, nutrient shortages, humidity, and other issues. The other portion
is the network through which the technology transfers, normally the Internet. This is commonly
mentioned to as the “Internet of Things” (IoT) —networks of things that interconnect with other objects
and with computers over and done with the Internet. Ingale et al. (2016) [27] suggested a large amount
of data set will improve the analysis and predictive decision making power of actors related to the
agriculture sector. An agriculturist could take a photograph of a yield with his smartphone and exchange
it to a database where authority could assess the advancement of the item in light of its shading and
distinctive properties. Ribarics, P. (2016) [28] looks at the role big data plays in agriculture. Big Data
may form the basis for several new capabilities, including the identification of associations between
field, weather and commodity data for optimum irrigation, reproduction, optimum livestock feeding
and marketing. Prognostic analytics can be used to predict seed demand, nourishments, and animal
feedstuff and allow the supplier of agri-business to receipts appropriate actions to meet claim progress.
New pricing systems may be set up to help balance the market in line with the supply available. He
concludes that big data analytics is one of the solutions for sustainable farming and minimizing its
footprint. Carbonell, I. M. (2016) [29] presents predictive farming suggested using current data from
the sensors, historical data available in databases, and future events. It helps farming companies to
predict customer land requirements based on data and use data analytics. It explains how business
people collect data from farmers from an ethical point of view for predictive analysis and effect on the
confidentiality and safety of farmer’s data. Nandyala, C. S., & Kim, H. K. (2016) [30] recommend the
type of sensor methods, technologies, implementations and benefits of numerous kinds of sensors for
mobile services to be used in their decision making. To end with, it presents U-Agriculture Mobile
Services design centred on Sensor Cloud Infrastructure that not only benefits plantations and
solicitations, but service suppliers and societies also manage B&M data.
N. Nalini et al. (2017) [8] explains Data Analytics, Stuff Internet, and Health Care Studies. Upgrades
in agriculture through the use of the Internet of Things and sensor data analytics gathered field level
information. The application proposed here performs three main things. One is a compilation of
information, the second is identifying unknown occurrences and the third is applying results to the
application concerned. The data are from the sensors for agriculture analytics. These findings help us
predict the plants of the future and contribute to bearable agriculture. His proposed application for Agri-
IoT includes 6 main layers, as shown in figure 1.
End Operator
Solicitation
Tunneling of data
Communication rules
Hardware device
First, we chose the correct hardware to add to the problem statements to build Agri-IoT Product. Then,
there will be the data collection and cleaning process involved. Must be mined on the actually collected
data information and the application will extant the extracted results to end-users. The Agri-IoT can be
applied to seed collection, field mechanization, precision and accuracy planting, soil erosion, promotion
of agriculture and irrigation systems. This application’s problem is spatial imbalance and temporal
imbalance. The application's analytical section varies according to the area. The application's analytics
part produces the data, and we make the decision based on the report. So it should produce the report
within a short time. Kamilaris et al. (2017) [31] presents smart farms with the objective of developing
information technology in e-commerce and addressing the problem of farms including growing yields,
water conservation and soil and plant health, and promoting new sustainable agricultural science. With
a new open-source, IT framework named UCSB Smart farm, he achieved that goal. The smart farm is
a hybrid cloud platform designed to allow smallholder farmers and other farm professionals, researchers
and students to use analytics to improve environmental sustainability and food production efficiencies.
Nayyar, A., & Puri, V. (2017) [32] proposed IoT Smart Stick that will allow agriculturalists to have
data on live soil humidity, atmosphere temperature at a very little price tag, so live checking can be
prepared. Adequate water is essential for carrying out agricultural activities in an efficient manner.
Agriculture IoT is combined with Web Map Service (WMS) and Sensor Observation Service (SOS) to
guarantee appropriate irrigation water controlling and diminish water surplus in effect. Agriculture
Temperature and Moisture Control Smart IoT Based Agriculture Stick is being proposed. Nuvvula et
al. (2017) [33] introduced a system designed using Arduino expansion kit which links to light sensor
gauging the light intensity, atmosphere temperature/moisture sensor for receiving the humidity in the
environments, soil moistness sensor for measuring level of water, and air poisonousness measuring
carbon dioxide and oxygen stages. Device has been explicitly premeditated to analyze the atmosphere
and report to the agriculturalist, round-the-clock, via the Wi-Fi connectivity incorporated into it.
Marjani et al. (2017) [34] analyzed the connection between big data analytics and IoT. It proposes
architecture for big IoT data analytics and technologies for data mining. Instantaneous analytics is
characteristically achieved on data composed from sensors. Rapid data analytics techniques must
achieve an analytical consequence within a tiny period. It presents big data analytics approaches under
organization, bunching, connotation rule mining, and prediction categories. In smart agriculture, text
and image data are captured from IoT devices, and Hadoop application is used for big data analytics.
Maksimović et al. (2017) [35] adopt the application of green nanomaterial in recent farming to decline
the practices of injurious and expensive nourishments. Thus, the solicitation of nanotechnologies and
IoT in this area can seriously explore the difficult of sustained farming. It exemplifies a learning of
Green-IoT and Green nanotechnology’s part in understanding smart and supportable agriculture.
Bhavani et al. (2017) [36] reviewed tools and applications that support the execution of big data
applications for agriculture exercise. The comprehensive building of Hadoop Ecosystem and its
mechanisms has been scheduled along with their context. These apparatuses are used to forecast
upcoming analysis by using the calm Dataset. Rajeswari et al. (2017) [37] reviewed the assimilation of
various big data analytic techniques and implementation of it in various agricultural fields. The
challenges of big data analytics are discussed here. Different data analysis techniques such as prognostic
analytics, reference system, and data mining, time series analytic technique are applied to a different
area of agriculture. Gill et al. (2017) [38] Refers the Electronic Farm Record (EFR) which contains
earth temperatures plots and evidence, rainfall maps and information, electrical conductivity maps and
material, mugginess gratified statistics, air permeability maps, pH level statistics, historical progress
proceedings, security and crop connected data and online post comprising chirps, websites, new
withstands and articles in agribusiness diary.
Al-Kahtani, M. S. and L. Karim (2018) [39] presents the dynamic data processing method for big data
sensors. It suggested how to reduce energy consumption using a multi-tier architectural system, and
how to perform data aggregation in each layer to reduce the IoT device's energy consumption. Parent
nodes accumulate data as they are communicated at the origin of the tree on the way to the sink. The
network structure is primarily established on an in-link approach to data aggregation. When an event is
observed by a sensor, if the data is gained from a fixed sensor application the sensor implements an in-
network data combination scheme. When data comes from a mobile sensor device, a grid-based method
is used to integrate the data.
Layer 3 data aggregation (Response center –
SQL server)
The proposed Big Data accumulation and cleaning architecture works in three covers, as seen in Figure
2. The lower layer aggregates sensor data. The central layer group’s data at the base station and the
upper layer totals data in a dispersed mode on the big data server. The lower layer of this model contains
sensors that need to be configured in such a way that every event happens, matches the data with the
field value and then transfers it to the middle layer. More data aggregation occurs at the middle layer
and the top layer of the model receives partly organized data. This way reduces the difficulty of the top
layer data analysis. They conclude in this application that if the distance between the nodes (sensors or
devices) is lower than the energy consumption is lower, too. The sensor transmits the data from itself
to the sensor gateway in the shortest possible way. Wolfert et al. (2018) [40] presented a paper on Big
Data for Smart Agriculture. They find that the rapid Internet of Things and web computing growth
propels the indication of smart farming. This model summarizes the concept of smart farming as a
cyber-physical system along the management cycle, which points to intellectual expedients-associated
to the web-control the farm system. Smart devices encompass traditional resources (e.g., flood measure,
tractor, and notepad) by incorporating independent situation-consciousness through all sorts of sensors,
built-in cleverness, capable of performing or doing something the slightest bit, independent activities.
In this context, the business practices (lower layer) concentrate on the generation and utilization of big
data in agricultural process. This part is divided into the chain of data, and farm supervision and
agricultural processes. The data shackle cooperates with farm procedures and farm organization
procedures over and done with many processes of assessment making, in which facts theaters a
significant character. The link of investors (middle layer) includes all investors intricate in these
procedures, not just big data consumers but also corporations specializing in data processing and
controlling and program actors. The web management (upper layer) technology component focuses on
information set-up, which supports the chain of data. He concludes that sensor implementation and
analytics, prognostic modeling, and the use of improved models to handle yield failure danger and
improve feedstuff performance in livestock production are opportunities for big data applications in
food production. Big data are intended to deliver prognostic visions into upcoming agricultural likely
results (prognostic harvest model, prognostic nourish consumption model, etc.), drive operational
decisions in real time and reinvent corporate developments for faster, advanced action and changing
corporate models.
Liu et al.(2018) [41] propose an idea to syndicate smart agriculture and clean energy consumption by
using surplus clean energy in agriculture production. It uses smart agriculture to provision power system
with clean energy consumption. Here, smart farming and clean energy systems combine to meet power
demand. The bottleneck technologies for smart agriculture and clean energy systems are multiline scale
coupling, measurement and monitoring system, multisystem correlation-coupling system, economic
evaluation system and model of smart agriculture load. Aliev et al. (2018) [42] Presents a real-world
method to attain data on heat, moisture and soil moisture of plant life. It develops a model device and
android application to get bodily data and directs it to the cloud. Also presents the use of an ANN as a
time series predictor to forecast the temperature. They realize this over industrialized IoP device and
Android platform, which delivers the data, attained from the farming atmosphere to the web over the
Thing-speak platform. The Thing-Speak-IoT platform offers applications that contract you analyze and
posturize your data in MATLAB platform. Data from sensors can be transmitted to Thing-Speak from
other hardware. Jerzak et al. (2018) [43] examine the impending for a new Information and
Communication Technologies-led model of progression and expansion in the agronomic sector in India.
The project’s objective is too improved realize the role of ICTs in India’s future monetary development
and to make references for India to carry on to leapfrog the progress process using ICTs in key segments
including agriculture, among others. Ngo et al. (2018) [44] designed the system and database scheme
to implement continental level data warehouse that supports big data analytics and storage. It built an
efficient system to handle agriculture and practical data mining techniques. This model mainly contains
a raw data module, abstraction renovation module, incorporated information module and data mining
module. The data are usually extracted in the practice of a data cube beforehand it is analyzed in the
data mining segment. A data cube is a data building that permits reckless analysis of data conferring to
the numerous extents that outline a commercial problem. The data cubes are created by an online
analytical processing engine (OLAP). Aher et al. (2018) [45] propose fog-centric power effective
design for a smart agriculture system. Here the fog bulges are placed very close to land and real-time
data manipulations is processed in the fog layer, which lowers the work load in the cloud layer. Here
the fog layer collects data from agriculture IoT sensor layer and performs real-time processing in which
farmer gets a faster response. Cloud Layer handles the agriculture data received from the fog layer for
storage and processing using data centers located on the cloud. CobleKaur et al. (2018) [46] suggested
the complex world of the technological era, there is a unique and kind role for data analytics and applied
economics. Applied economics is equipped with unstructured data, and very relaxed. Agriculture and
applied economics read the next generation with geospatial analytical preparation and techniques.
Kovács, I., & Host, I. (2018) [47], suggested document 4.0 on the part of digitization in agriculture.
This paper systematizes the art of smart farming, the IoT, cloud computing and big data. Geo-location
tools such as GPS, GIS, crop tracing, precision earth sampling, proximal and distant recognising,
without human in-flight vehicles, auto-managed and guided apparatus, and flexible rate equipment
allow field-based operations. Animal-attentive technologies include RFID chips and automated milking
and feeding systems, among others. The StarFire 3000 receiver collects satellite broadcasting signals
from the GPS and can use GLONASS satellites to preserve supervision recital even under sheltered
circumstances and other changeable situations. Thus, the receiver should compensate in the field to
ensure true vehicle location regarding the ground. Finally, he concludes that smart farming uses
technology, sensors and big data from GPS systems, device to device (M2M) and the Internet of Things
to maximize crop yields and minimize waste. Kumar, H., & Menakadevi, T. (2018) [48], published a
review paper on data analytics in the agricultural sector. This paper discusses Big Data applications to
support agriculture and methods to help the execution of Big Data applications for agricultural facilities.
He noted that all data collected were mapped to the organization system using a topographical
information system and advanced sensor apparatus, which will improve the crop’s sustainable
productivity with greater benefit. This paper also includes the implementation of the Apache Hadoop
framework and the core elements thereof. Ultimately, he concludes that precision farming provides
farmers with technological backup support for applying technology in agriculture. In addition, it also
found that Hadoop is the best platform for processing massive data sets in agriculture. B. Santhiya et
al. (2018) [49] suggests using the LPC2418 microcontroller to direct and obtain data from the web-
based network using web-enabled applications. It had QOS including good response time, productivity
and availability that would be ideally suited for monitoring and managing townhouse gardens from
remote locations.
Sarker et al. (2019) [50] focuses on the suitability of big data technologies in the agriculture sector. The
big data based digital agricultural technology can help farmers from the field preparation to harvesting,
such as weather forecasting, crop yield prediction, yield collection, irrigation supervision, crop viruses
and pest supervision, agricultural selling and agricultural pest management. Jain, A., & Kumar, A.
(2019) [51] Presents a model that collects temperature data and water level data from various sensors
and displays it in an LCD screen. Then, these data are given to microcontroller, which analyses it and
transmits it to the server using Wi-Fi connection. The server performs the data analytics on data and
generates the result in the form of output. Mathivanan, S., & Jayagopal, P. (2019) [52] provides a
complete evaluation of big data virtualization in the agriculture purview. The big data virtualization
enables farmers to take well decision-making tools such as SQL virtualization for integrated access,
query, reportage, prognostic analytics, back-end data warehouse, such as Hadoop, and NoSQL. Due to
the accumulative complexity of big data, we need data virtualization, and its arrangement. Surya, P., &
Laurence Aroquiaraj, I. (2019) [53], proposed an overview of efficiency of K-means and K-medoid
clustering algorithms using data set on agriculture. Data mining is a method of mining information from
a massive sized data set, where clustering is used to identify related items within the dataset. The main
goal of this paper is to find the best clustering technique in the Agriculture dataset for classifying a
similar group of objects. In an article he used the clustering technique for evaluating the report on
agriculture. He applied clustering techniques for K-means and K-medoid to the Agriculture database.
Cluster is a data object array and clustering is the technique for partitioning or segmenting the data into
classes. The most associated data are grouped into the cluster. The accuracy, precision, and recall in the
case of the k-means algorithm are poor after the experimental analysis, and it is high in the case of k-
medoid. The mean square error and square root mean error are high in k-means, and small in k-medoid.
Results exhibits the k-medoid clustering algorithm has great precision and a lesser error rate compared
to the k-means clustering algorithm. By these findings, k-medoids gave this Agriculture dataset a better
output. Anushree et al. (2019) [54] proposed an article on the evaluation of a network of wireless sensors
in precision agriculture. This paper consists of a survey on wireless sensor networks on Precision
agriculture, with different web-based IOT systems. A web-based IoT architecture is used that consists
of three layers. The front-end layer applies the requisite agricultural activities that are collected as
information about the environment. The gateway layer links the front-end layer to the internet and the
entire process that is occurring, and the data storage in the back-end layer has happened. He clarified
the design of WSN and of WSN nodes. He concludes that many agricultural production systems will
increase the efficiency, productivity and profitability of precision agriculture by using Wireless Sensor
Networks (WSNs). Liu et al. (2019) [55] proposed a solution for agricultural issues using modern
technologies IoT and big data solutions. It uses lot of vulnerable source application and open library
documents. For access hardware, it uses ports, RPI.GPIO and WiringPi. PySerial is used for UART
ports. Video scrutiny is useful and constructed on motion. Paho-python is used for communication
broadcast and analysis, web service is made with WebIOPi. Other components delivered by Python are
also functional. Through video reconnaissance, operators identify the development of land activity
remotely and in real-time and monitor the actual process rank of the Internet of Things device. One of
the basic purposes of the model is to deliver instantaneous data requests and information. For that, it
uses Hadoop service EMR (Elastic MapReduce) by Amazon. Through this, it is possible to rapidly
conduct analysis through machine learning, confirm the expectation of demonstrating, and then use the
system to forecast on a huge gauge. Jboss is chosen as the development server of the system, which is
openly available and with free of cost, and has a less source employment, and wires highest level of
concurrency. The client side of web server is Nginx to distinct the static and dynamic resources. Chen,
J., & Yang, A. (2019) [56] makes a smart agricultural system is made using IoT, data imagining analysis
and constellation analysis. Agricultural activities are promoted by detecting, identification, broadcast,
checking and feedback of IoT. The collection of agricultural data has done by IoT and data analyses
were performed by visual interactive system. IoT mainly has four core technologies that are sensor
equipment, RFID equipment, and rapid response code and embedded system technology. The
visualization technology is used to collect data, which is transferred to the cloud where cluster
processing was achieved and placed in the data source. Then, the information is transferred to the user
terminal in the form of a graph. The spring framework is used for data visualization, which contains a
web browser, web server and SQL database. The main responsibility of a web browser is to interact,
parse, map and display modules. Web server interacts with user requests, querying databases and,
fetches relevant data from the database. SQL database is used to store relevant data. Kamath et al.
(2019) [57] analyze the execution of a wire free chromatic sensor network for accuracy farming. In
sensor nodes, Bluetooth 4.0 directs data for the base station. Bottom station forwards data to the distant
station using IEEE 802.11 a/b/g/n standard. Planetary cubicle battery used as the control source at both
base station and sensor nodes. By using support vector machine classifiers different shape features were
extracted at the remote station. The bottom station receives the images and stores them in the line
maintained by RabbitMQ. Plant objects from images developed by Raspberry Pi. A agent that performs
a random forest classifier, and evaluation type of classification. From this application, we observed that
a wireless ocular sensor network could be established for checking crops for pests using Raspberry Pi.
The extended this network by using different sensors such as earth humidity sensors, optical sensors,
moisture sensors, and temperature sensors that a low-cost, full-fledged yield-checking model is
developed. Hu et al. (2019) [58], introduced relay-aided Non-Orthogonal Multiple Access (NOMA) for
uplink transmission in WSNs to achieve smart and intelligent agriculture. This NOMA technique can
lodge extra sensor nodes with the same Relay Equipment’s (RE).NOMA can transmit several symbols
at the same time on the same Relay Equipment’s (RE’s) by excruciating them with power field and
discriminate them with different power levels. The next generation communication technology 5G,
NOMA achieved a high data transfer rate and huge uplink and downlink transmissions. Tseng et al.
(2019) [59] introduced an intelligent agriculture platform using IoT and data analysis to monitor
environmental factor and yield analysis using 3D cluster analysis. In this scheme, data normalization is
attained using a mixture of much average variance. The 3D analysis is used to analyze the environmental
factor. The System arrangements the vital value in the selected group based on the prospect environment
and offers guidance to the former whether the crop is appropriate to farm. This intelligent agriculture
has futures such as IoT sensors, solar storage system and web platforms that help assemble evidence
from IoT sensors. Big data analysis approach includes data normalization using average variance, 3D
analysis to analyze the connection between different ecological factors, to determine whether the
selected crop belongs in the suitable group, and setting significant value in the group for giving advice
on former about the crop cultivated. Farooq et al.(2019) [60] presented core technologies of smart
farming that was network technology used in IoT, cloud computing, big data storage and analytics. The
4 major mechanisms are the physical construction, data gaining, data dispensation, and data analytics.
The physical structure is designed and it controls devices. Data acquirement is further divided into IoT
data acquirement and normal data acquirement. Data processing includes picture and video
dispensation, data loading, decision carry system, and data mining. Finally, data analytics has 2 main
features include monitoring and controlling. It also analyzed the different communication protocols of
network technology, including WiFi, Lora Wan and code division multiple access (CDMA) technology.
ZigBee network is the main Internet technology used for long distance. Tóth, M., Felföldi, J., &
Szilágyi, R. (2019) [61] explains how IoT technology was analysed and tested with sensor technology
and wireless network integration system. Remote Monitoring System (RMS) is proposed to incorporate
an approach with web and wireless communications. The key aim is to gather timely agrarian
production atmosphere data that enable easy admittance to agricultural amenities such as warnings via
the SMS and weather conditions, crops. Rao et al. (2019) [62] study large-scale farming using IoT and
data analytics. Here he overcomes the issue of rising power consumption in IoT devices through the
implementation of a centralized sensor control unit. A centralized platform is deployed with a Zigbee-
connected wireless sensor network. He argues that Zigbee has greater wireless connectivity than Wi-Fi
connectivity.
Rasooli et al. (2020) [63] suggested the development of a system called the optimal watering of crops
based on a WSN. This research was designed at scheming and evolving a system that uses harvest
ground node sensors with data supervision over mobile and an online application. The three elements
are the Software, Internet, and Phone. Pini et al. (2020) [64] presented a trial setup and practice to test
the performance of RTK-GNSS based devices in the functional demonstration. Also, extracts the limits
of GNSS in precision agriculture based on small robots and AI. The analysis confirms that RTK
provides the best performance and maximum level of location accuracy and exactness for agricultural
talks. The RTKGNSS based device matches a performance requirement which is used for machinery
direction and involuntary field management using a wireless network connection. Saad, A., & Gamatié,
A. (2020) [65] presents recent works on management of water and agricultural work miniaturization
system by using advanced technologies. By simplifying water management practice by using
appropriate computerization level and provides agriculturist to connect to the system at any time
anywhere basis. The sensors are installed in WSN and CPS setup and it produce enormous data stored
and handled in a large scale. CPS is an assembly of wireless sensor device with low-level calculating,
storage of data and message abilities. The system alienated into multiple layers cooperating with
unswervingly by using a wireless network frontend and backend. The foremost layer consists of IoT
devices and the next layer contains of storage of data and servers itself makes cloud environment. APIs
can be delivered to recover data in the web platform or expansion of mobile applications. Gupta et al.
(2020) [66] provide a summary of smart agriculture focused on multilayered structure. The proposed
architecture exposes IoT and Cyber Physical System (CPS) multilayer architecture. The cloud and edge
services are capable of storing and process large amount of data generated by physical devices. It also
considers, massive amounts of data composed at the edge or cloud layers, and climaxes the need for
numerous multi-cloud or edge-cloud scenarios. The distinctive construction contains mainly 4 layers
such as corporal layer, edge layer, fog layer and network message layer. Liu et al. (2020) [67] proposed
an exactness directive model of ACPS. This model interacts with the environment to improve water and
fertilizer regulations. It verifies and analyzes PRMWFA-ACPS by relating the replicated and dignified
values of development period, leaf characteristic, and soil moisture content. This model built an
inclusive biophysical model into ACPS, which comprises construction of a biophysical model in ACPS
that interrelates with the communication model, constructing a model of water and fertilizer for alfalfa
based in ACPS (PRMWFA-ACPS). The PRMWFA-ACPS mainly comprises 3 gears such as physical
atmosphere, calculation situation and cyber-physical communication. The physical atmosphere
involves of environment conditions such as temperature, solar radiation and basic ecological limitations.
The computation atmosphere is composed of calculation entities, which are accuracy rule approaches.
Gulec, O., Haytaoglu, E., & Tokat, S. (2020) [68] introduced a newly distributed CDS algorithm namely
CDSSEHA, in WSN, solar energy harvester node, especially for accuracy agriculture application. The
various types of corporal, substance and natural features of agriculture are deliberate and used in smart
agricultural applications. Here application uses mobile sensors that move randomly on the farm and
gathers the data. These data were analyzed in base station to protect against bacterial and fungal disease
through an image processing technique. The communication backbone was constructed between node
and base station was achieved using CDSSEHA. The performance of the proposed application is
achieved through the said algorithm through the network. Huang et al. (2020) [69] Introduce a new
concept called PAIoT. This system uses renewable power in agriculture that was controlled by the IOT
device. The multimedia data stream of disease and creature pests, crop progress, broadcast and soil
moistness and procedure status of the photovoltaic module is gained during the process. The
photovoltaic panel is complete use of water property, node installation and optimization of expenditure
for sensor networks through multiple functions. Optimize image transmission and power generation on
the environmental climate. The optimization of image data transmission can be achieved using 3
methods such as communication quality, image quality, storage and computational capability.
Computational capability can be achieved by edge computing, cloud computing, and fog computing
that supports big data processes.
Nalini N. et al. Data Sensor gathered information Seed choice, farm automation,
(2017) [8] Analytics, from field level. Finding precision agriculture, soil
IOT unknown events and apply corrosion, Irrigation system.
findings to relevant
application.
Nayyar, A., & IOT, Cloud IOT based smart farming and Provide farmers with more than
Puri, V. (2017) computing precision agriculture. 99 percent accurate results to help
[32] them get reliable live feed from
ambient temperature and soil
moisture.
Nuvvula et al. IoT Hydroponics, Controlled The device can comparatively
(2017) [33] Environment Agriculture simple to operate by the end
(CEA) users, and can be applied in
small as well as large scale
agribusiness.
Marjani et al. Big data, data Classification, clustering. Get moistness level of soil, stem
(2017) [34] analytics diameter of plants, micro
environment circumstances.
Maksimović et IoT and nano Green nano technology Making the farming segment and
al. (2017) [35] technology food business more proficient,
harmless and viable
Bhavani et al. Big data Precision Agriculture Deliver technical holdup support
(2017) [36] analytics techniques to the agriculturalists to
instrument advanced models that
are simulated in a large scale.
Rajeswari et Big data Predictive analytics, Enhance productivity and reduce
al. (2017) [37] analytics Recommendation System, manual inputs.
Data Mining, Time series
analytic technique using Big
data
Gill et al. Big data Global Positioning System Accuracy agribusiness.
(2017) [38] analytics (GPS), Yield monitoring and
mapping, Information
management, Variable-rate
Technology (VRT)
Al-kahtani, M. Data Data aggregation at layer It reduces the complexity of data
S., &Karim, L. aggregation 1(Sensor network), layer 2 analysis at the top layer.
(2018) [39] (Base station – Internet), layer
3(Response center – SQL
server)
Wolfert, S., Big data Predictive yield model. Accomplish feed throughput in
Ge, L., et al livestock production, drive
(2018) [40] working decisions in real time
and redesign business processes
for quicker, creative action and
commercial models that change
the game.
Liu, J., Chai, Data analytics Photovoltaic systems, Improves agriculture yield
Y., et al. bottlenecks and related production using clean energy
(2018) [41] technologies for coupling model.
agriculture and clean energy
systems, covering multi-time
scale coupling, multi-system
correlation coupling
mechanisms
Aliev et al. IoT, WSN nonlinear autoregressive Proficient to monitor ambient
(2018) [42] model, novel WSN approach, temperature, moistness and soil
predictive model moisture.
Jerzak et al. Information ICT enabled agriculture To well apprehend the role of
(2018) [43] Technology ICTs in India’s future financial
development
Ngo et al. Data analysis, crop yield prediction, System to efficiently handle
(2018) [44] Data precision agriculture agricultural data Efficiently
warehouse forecast and improve crop yield.
Aher et al. Cloud Fog computing, real time Gives faster response in real time
(2018) [45] computing processing applications and also reduces the
load on the cloud, less network
cost and reduced energy
consumption.
Coble, K. H., Big data Precision agriculture, Climate forecasting, crop yield
et al (2018) Analytical method, applied and crop selection prediction,
[46] economics. irrigation systems, prediction of
crop disease, agricultural policy
and trade.
Kovács, I., & IOT, Cloud GPS, GIS Optimize crop yield and reduces
Husti, I. computing, wastage.
(2018) [47] Big data
Kumar M, H., Data analytics Apache Hadoop Framework Precision farming provides
& and its core elements. farmers with technical support for
Menakadevi, implementing technology in
T. (2018) [48] agriculture. And also found that
Hadoop is ideally suited to
processing large-scale data set for
agriculture.
B. Santhiya IOT Web based automatic control Automates irrigation and
(2018) [49] Cloud system. monitoring of plants in the terrace
computing garden
Sarker et al. Big data and Digital agriculture Help farmers from the field
(2019) [50] analytics preparation to harvestings such
as weather forecasting, crop
yield prediction, agricultural
marketing and agricultural pest
management.
Jain, A., & IoT Sensor based data gathering Reduce costs and stimulate
Kumar, A. and processing normal growth productivity,
(2019) [51] profitable and warm repairs.
Mathivanan, big data Big data virtualization Plummeting the danger of data
S., & mistake, capacity, progress, data
Jayagopal, P. storage, policies and also an
(2019) [52] increase in the speed of data
access
Surya, P., Data mining K-means and K-medoid K-medoid clustering algorithm
&Laurence clustering algorithms. achieved satisfactorily high
Aroquiaraj, I. accuracy and lower error rate
(2019) [53] compared to the clustering
algorithm k-means. By these
results k-medoids gave this
Chen, J., & IoT, data Data visualization approach, Most relevant data was obtained
Yang, A. visualization Cluster analysis approach by front end visualization
(2019) [56] analysis and technology.
cluster
analysis.
Kamath, R., WSN, Decision support system Bluetooth 4.0 can be used as
Balachandra, Data approach, message technology,
M., et al. classification wireless visual sensor network Data sufferers can be condensed;
(2019) [57] approach, The excellence of images made is
Bluetooth 4.0, very subtle and brittle well suited
Data classification. for outside location.
Hu, Z., Xu, L., WSN, IoT Non Orthogonal Multiple Supports more user equipment’s
Cao, L., et al. Access (NOMA) simultaneously.
(2019) [58] Supports high data transmission
rate and massive connectivity.
Tseng, F. H., IoT, Data Data cleaning and Analyzes cultivation techniques
Cho, H. H., & analytics normalization, 3D cluster practiced by customer, examines
Wu, H. T. correlation analysis, Crop environmental changes, choose
(2019) [59] selection analysis and suitable crop for cultivation.
decision making.
Farooq, M. S., IoT, Data Cloud computing, Big dataNetwork architecture, platform
Riaz, S., Abid, Mining, Data storage and analytics. and design helps access to IoT,
A., et al. analytics improves crop productivity,
(2019) [60], Provides overview of IoT
applications, sensors, protocols
and data enabled technologies.
Tóth, M., sensor WSN, Remote Monitoring Collecting real-time farm
Felföldi, J., & technology, System (RMS) construction environment data.
Szilágyi, R.wireless
(2019) [61] network,
IOT
Rao, G. P. R., IOT and data Centralized control unit for Minimizes power consumption
Indira, V. V. analytics sensors. for devices with IOT.
S., et al.
(2019) [62]
Rasooli, M. WSN, data node sensors and web Optimal irrigation of the water for
W., Bhushan, analytics application farming crops.
B., & Kumar,
N. (2020) [63]
Pini, M., Wireless Global navigation satellite Provide the maximum level of
Marucco, G., sensor system (GNSS), real time position correctness (cm range)
et al. (2020) network, data kinematic (RTK) and precision for agricultural
[64] analytics tasks.
Saad, A., & Wireless Renewable energy, Machine It discourses usage of water in
Gamatié, A. Sensor learning using K-Means agriculture.
(2020) [65] Networks clustering. Web and mobile
(WSN), based applications, motion
Cyber- detection devices
Physical
System (CPS),
Cloud
Computing
Gupta, M., IoT, cloud Cloud based real time data Elaborates cyber security
Abdelsalam, computing, gathering and processing. challenges in smart farming.
M., et al. Cyber Physical
(2020) [66] System (CPS)
Liu, R., Zhang, Cyber Physical precision regulation model, It improves water and fertilizer
Y., Ge, Y., Hu, System (CPS), agriculture cyber physical regulations.
W., & Sha, B. Cloud system (ACPS), PRMWFA-
(2020) [67] computing ACPS MODEL:
Gulec, O., WSN, Solar Image processing technique. Prevents data loss and collusion,
Haytaoglu, E., energy system. increases the lifetime of WSN.
& Tokat, S.
(2020) [68]
Huang, K., Photovoltaic PAIoT Solves the problem of an energy
Shu, L., Li, K., agriculture. shortage, Use of renewable
Yang, F., et al. IoT, Big data energy sources.
(2020) [69]
Balamurugan., big data Supervised and no supervised Better decision support system,
et al. (2020) analytics learning models, Linear water and other resource
[70] regression, Map Reduce management.
Algorithms
Field level data are soil data, insecticide data, external data, and environmental data, as shown in figure
3. Soil data give soil type, soil water content, soil fertilizer content, etc. Data on insecticides or
pesticides provide information on how much crop was affected by the pesticide. External data provide
information on animal or bird attacks on the field. Environmental data include data on temperature,
humidity, and rain data and so on. The data on the price of the crop, the price of the agricultural machine,
the price of farm equipment and market price forecast for the next year as shown in figure 4 are given
in market level data. The problems related to data collection and analyses occur here. Different sensors
and wireless sensor networks need to be used to collect data from the field level. To access business
data, the current data collection must be used from the cloud [72]. Both these field level data accessed
by sensors and market level data accessed from database are essential data that constitute big data that
plays a major role in smart agriculture system.
Our proposed smart agriculture system architecture has 5 layers, as shown in Figure 5. Here, the field
level sensors are the topmost layer, and the user interface is the bottommost layer. Here field-level
sensors such as temperature sensors, soil sensors, water level sensors, moisture sensors, a rain detector
sensor, and object detector sensors are mounted in the field performing their role. A device called the
Sensor Control System (SCS) controls all of these different sensors. This SCS supplies various sensors
with the necessary power and collects data from it using the Wireless Sensor Network (WSN) [73]. All
of these data collected from various sensors will be stored via the internet in a cloud environment. Such
field-level and business data will then be processed and displayed in the UI layer.
8. IMPLEMENTATION :
Within the smart agriculture architecture at what point, we must see which technique is best for
implementation in each layer. The numerous field-level sensors such as temperature sensors, water level
sensors, moisture sensors, soil sensors and object detector sensors are mounted on the field at different
positions starting from the top level. In the second level Sensor Control System (SCS) requires a system
board with ARM MCU, which acts as a system for data acquisition and control.
To read the file, a client API sends file pointer frame indexes to the Name node as a request. As a
result name node must address every data node that has the file replica. The application API
interacts with the name node for the granting of data nodes that serve as a primary to write a file
[87]. The client is updating its changes to all Data Nodes, but this apprises are stored in a distinct
Data Node buffer. After all the changes in Data Nodes, the client will then send a "Commit" appeal
to the primary which will decide an instruction to apprise all secondary Nodes. When secondary
nodes complete an operation commit, primary will send a response about its progress to the client.
MapReduce: It has two functions that are likely to combine and reduce function as seen in Figure
9. Once a client program submits task outline to the job tracker it begins execution. The job tracker’s
role is to pick multiple frames and task trackers based on its network [88]. The work tracker then
sends requests for assignments to the selected task tracker. The mission tracker collects and
separates data from inputs and performs a map processing function. When the work tracker is
complete it notifies the job tracker. Then the work tracker selects to alert any activity tracker to
perform the reduction function. Each task tracker reads the local file if crashes occur during the
map and reduce its assai function. Each task tracker reads the local files if any crashes happen
during the map and reduce function it assigns the task to another task tracker in the network. If both
the map and reduce phase complete its execution than only the job tracker will unblock the client
program [89].
the accuracy and privacy of the data during transmission. For this reason, a suitable network must
transmit data from the field node to the base station. A reliable wireless sensor network should be
established with various factors such as quality, privacy, availability and low power consumption, and
should be feasible to the farmer.
Research Gap 3: The proper data analytical techniques to solve real-time problems. The data gathered
from sensor nodes or from the cloud database are effective and meaningful only when it was properly
processed and analyzed using the proper technique. Depends on the type and nature of the data obtained
we must adopt appropriate data analysis techniques that should meet real-time requirements. So the
appropriate data analytics techniques will help farmers get specific information that will lead them to
take certain actions to further improve their agricultural process. Technology should not replace farmers
but it should assist them toward positive change.
Research Gap 4: Requires user interface design to display analyzed results. The main aim of the
system is to display the analyzed result in the proper user interface that supports all types of devices.
The entire system’s work is said to be successful if it can able to convey results in user understandable
format through a proper interface. So in smart agriculture system with a proper user interface is also a
key concept while developing the entire system.
12. RESEARCH AGENDA :
(1) What procedures and approaches were used to gather and process data from sensors and
databases?
(2) What framework is used to integrate IoT and data analytics to achieve a smart agriculture
system?
(3) What new technology can be proposed from in IoT and data analytics perspective to achieve
better results in a smart agricultural system?
(4) What Wireless Sensor Network (WSN) technology can be suggested for the agricultural sector
by considering software, hardware and network complexities?
(5) What farming strategies have evolved in the smart agriculture system, still continues to
improve?
(6) What are the challenges in the implementation of different technologies in the agricultural
process?
(7) What IoT and data analytics architecture recommend low cost, reliable technologies that make
more benefits for farmers in rural areas.
13. ANALYSIS OF RESEARCH AGENDAS :
The different procedures and approaches can be used to gather and process the data in a smart
agricultural system that provides expected results. We must select the appropriate technique that meets
all aspects of the social scene. The development of a framework to integrate both IoT and Data analytics
is to be developed with the best suitable existing hardware and software. There should be a need to
survey and analyze the existing current technology in both IoT and data analytics to achieve a better
result in smart agriculture. The research is mainly focused on selecting the best technology in WSN,
IoT and data analytics that gives solutions based on available data at low cost, reliable and should make
benefit farmers worldwide. The implementation of data analysis technology is challenging in
agriculture because the data are not the same for all crops. So based on the nature and type of data, one
should find a relevant method for data analytics. The farmer can adopt any strategies during the farming
process to achieve maximum profit. He can perform the farming process based on different parameters
such as soil conditions, weather conditions, water availability, and price of the commodity. So here
essential to provide this information well in advance to farmers to help them take the decision to perform
the farming process. The prediction of all this parametric information is the biggest challenge that needs
to be experiment in the later stage. The yield prediction and price prediction are useful for farmers
worldwide to perform the farming process without having the risk of a fall in yield and price. So the
fulfillment of this entire research agenda will help farmer’s community in the entire world and thus we
achieve good work by contributing to a large number of populations.
17. CONCLUSION :
A sector like agriculture should improve through the implementation of advanced technologies such as
the Internet of Things (IoT) and data analytics. It helps achieve precision farming that allows farmers
to get more accurate information on both on the field and off-field agriculture. The two main aspects of
agriculture are context awareness, and fault tolerance can be achieved. The availability of the most
accurate and effective sensors and the availability of internet facilities in all parts of the agricultural
land could rectify all of these. So I conclude that the population of the Indians and world is growing
exponentially, so the whole world is anticipating food, so we must grow more agricultural crops to meet
the requirements of the world by implementing advanced technologies in agricultural practice.
REFERENCES :
[1] Patil, V. C., Al-Gaadi, K. A., Biradar, D. P., & Rangaswamy, M. (2012). Internet of things (IoT)
and cloud computing for agriculture: An overview. Proceedings of agro-informatics and precision
agriculture (AIPA 2012), India, 292-296.
[2] Patil, K. A., & Kale, N. R. (2016, December). A model for smart agriculture using IoT. In 2016
International Conference on Global Trends in Signal Processing, Information Computing and
Communication (ICGTSPICC), 543-545.IEEE.
[3] Veena, S., Mahesh, K., Rajesh, M., & Salmon, S. (2018). The survey on smart agriculture using
IOT. Int J Innov Res EngManag (IJRIREM), 5(2), 63-66.
[4] Chauhan, N., Krishnakanth, M., Kumar, G. P., Jotwani, P., Tandon, U., Gosh, A., ...& Santhi, V.
(2019, March). Crop Shop–An application to maximize profit for farmers. In 2019 International
Conference on Vision Towards Emerging Trends in Communication and Networking (ViTECoN), 1-7.
IEEE.
[5] Barcelo-Ordinas, J. M., Chanet, J. P., Hou, K. M., &García-Vidal, J. (2013). A survey of wireless
sensor technologies applied to precision agriculture. In Precision agriculture’13, 801-808.Wageningen
Academic Publishers, Wageningen.
[6] Mallick, C., & Satpathy, S. (2018). Challenges and Design Goals of Wireless Sensor Networks: A
State-of-the-art Review. International Journal of Computer Applications, 179(28), 42-47.
[7] Hung, M. C., Wu, J., Chang, J. H., & Yang, D. L. (2005). An efficient k-means clustering algorithm
using simple partitioning. Journal of information science and engineering, 21(6), 1157-1177.
[8] Nalini, N., & Suvithavani, P. (2017). A Study on Data Analytics: Internet of Things &Health-
Care. International Journal of Computer Science and Mobile Computing (IJCSMC), 6(3), 20-27.
[9] Sawaitul, S. D., Wagh, K. P., & Chatur, P. N. (2012). Classification and prediction of future weather
by using back propagation algorithm-an approach. International Journal of Emerging Technology and
Advanced Engineering, 2(1), 110-113.
[10] Jagyasi, B., Mohite, J., & Pappula, S. (2013). Applications of mobile sensing technologies in
precision agriculture. CSI Communications, 18(1), 21-23.
[11] Li, D., Zheng, Y., & Zhao, W. (2019). Fault analysis system for agricultural machinery based on
big data. IEEE Access, 7, 99136-99151.
[12] Kang, H. H., Lee, Y. J., & Han, W. O. (2009). Energy-Efficient Hierarchical Cluster-Based Routing
for Ubiquitous Sensor Networks. The Journal of the Korea institute of electronic communication
sciences, 4(3), 243-246.
[13] Mahmood, T., &Afzal, U. (2013, December). Security analytics: Big data analytics for
cybersecurity: A review of trends, techniques and tools. In 2013 2nd national conference on
Information assurance (ncia), 129-134.IEEE.
[14] Yang, Y., & Chen, W. (2016). Taiga: performance optimization of the C4. 5 decision tree
construction algorithm. Tsinghua Science and Technology, 21(4), 415-425.
[15] Jeong, S., Jeong, H., Kim, H., & Yoe, H. (2013). Cloud computing based livestock monitoring and
disease forecasting system. International Journal of Smart Home, 7(6), 313-320.
[16] Medar, R. A., & Rajpurohit, V. S. (2014). A survey on data mining techniques for crop yield
prediction. International Journal of Advance Research in Computer Science and Management
Studies, 2(9), 59-64.
[17] Parthasarathy, N. I. D. V. (2017). Evaluation and Management of Soil Fertility. International
Journal of Pure and Applied Mathematics, 117(8), 11-15.
[18] Sankaralingam, B. P., & Sarangapani, U. (2016). Agronomic-meteorological model for weather
forecasting to predict the rainfall using machine learning techniques. International Journal of
Convergence Computing, 2(2), 183-192.
[19] Yadav, R., Rathod, J., & Nair, V. (2015). Big data meets small sensors in precision
agriculture. International Journal of Computer Applications, 975(1), 8887 - 8895.
[20] Forkan, A. R. M., Khalil, I., Ibaida, A., & Tari, Z. (2015). BDCaM: Big data for context-aware
monitoring—A personalized knowledge discovery framework for assisted healthcare. IEEE
transactions on cloud computing, 5(4), 628-641.
[21] Li, X., Eckert, M., Martinez, J. F., & Rubio, G. (2015). Context aware middleware architectures:
survey and challenges. Sensors, 15(8), 20570-20607.
[22] Channe, H., Kothari, S., & Kadam, D. (2015). Multidisciplinary model for smart agriculture using
internet-of-things (IoT), sensors, cloud-computing, mobile-computing & big-data analysis. Int. J.
Computer Technology & Applications, 6(3), 374-382.
[23] Sravanthi, K., & Reddy, T. S. (2015). Applications of big data in various fields. International
Journal of Computer Science and Information Technologies, 6(5), 4629-4632.
[24] Ramesh, D., & Vardhan, B. V. (2015). Analysis of crop yield prediction using data mining
techniques. International Journal of research in engineering and technology, 4(1), 47-473.
[25] Ravisankar, K., Sidhardha, K., & Prabadevi, B. (2017). Analysis of agricultural data using big data
analytics. Journal of Chemical and Pharmaceutical Sciences, 10(3), 1132-1135.
[26] Stubb, M. (2016). Big data in US agriculture. Washington, DC: Congressional Research Service.
[27] Ingale, V. B., & Jadhav, D. K. (2016). Big Data A Great Revolution in Precision Agriculture using
Predictive Weather Analysis and Soil Analysis. International Journal of Agriculture Innovations and
Research, 5(3), 410-412.
[28] Ribarics, P. (2016). Big Data and its impact on agriculture. Ecocycles, 2(1), 33-34.
[29] Carbonell, I. (2016). The ethics of big data in big agriculture. Internet Policy Review, 5(1), 31-39.
[30] Nandyala, C. S., & Kim, H. K. (2016). Big and meta data management for U-agriculture mobile
services. Int. J. Software Eng. Appl. (IJSEIA), 10(1), 257-270.
[31] Kamilaris, A., Kartakoullis, A., & Prenafeta-Boldú, F. X. (2017). A review on the practice of big
data analysis in agriculture. Computers and Electronics in Agriculture, 143(1), 23-37.
[32] Nayyar, A., & Puri, V. (2016, September). Smart farming: IoT based smart sensors agriculture
stick for live temperature and moisture monitoring using Arduino, cloud computing & solar technology.
In Proc. of The International Conference on Communication and Computing Systems (ICCCS-
2016) (pp. 9781315364094-121).
[33] Nuvvula, J., Adiraju, S., Mubin, S., Shahana, B., & Valisetty, V. (2017). Environmental Smart
Agriculture Monitoring System Using Internet of Things. International Journal of Pure and Applied
Mathematics, 115(6), 313-320.
[34] Marjani, M., Nasaruddin, F., Gani, A., Karim, A., Hashem, I. A. T., Siddiqa, A., & Yaqoob, I.
(2017). Big IoT data analytics: architecture, opportunities, and open research challenges. IEEE
Access, 5, 5247-5261.
[35] Maksimović, M., & Omanović-Mikličanin, E. (2017). Green internet of things and green
nanotechnology role in realizing smart and sustainable agriculture. In VIII international scientific
agriculture symposium “AGROSYM 2017, 2290-2295.
[36] Bhavani, D. D., & Bharati, R. B. S. (2017). An Efficient Method to Incorporate Precision Farming
in Indian Agriculture Using Robotics and Internet of Things. International Journal of Research in IT &
Management, 6(9), 71-79.
[37] Rajeswari, S., Suthendran, K., & Rajakumar, K. (2017, June). A smart agricultural model by
integrating IoT, mobile and cloud-based big data analytics. In 2017 international conference on
intelligent computing and control (I2C2), 1-5. IEEE.
[38] Gill, S. S., Chana, I., & Buyya, R. (2017). IoT based agriculture as a cloud and big data service:
the beginning of digital India. Journal of Organizational and End User Computing (JOEUC), 29(4), 1-
23.
[39] Al-Kahtani, M. S., & Karim, L. (2018). Dynamic data aggregation approach for sensor-based big
data. International Journal of Advanced Computer Science and Applications (IJACSA), 9(7), 62-72.
[40] Wolfert, S., Ge, L., Verdouw, C., & Bogaardt, M. J. (2017). Big data in smart farming–a
review. Agricultural Systems, 153(1), 69-80.
[41] Liu, J., Chai, Y., Xiang, Y., Zhang, X., Gou, S., & Liu, Y. (2018). Clean energy consumption of
power systems towards smart agriculture: roadmap, bottlenecks and technologies. CSEE Journal of
Power and Energy Systems, 4(3), 273-282.
[42] Aliev, K., Pasero, E., Jawaid, M. M., Narejo, S., & Pulatov, A. (2018). Internet of plants application
for smart agriculture. Int J AdvComputSciAppl, 9(4), 421-429.
[43] Jerzak, M., Smiglak-Krajewska, M., Czerwinska-Kayzer, D., & Florek, J. (2018). The market for
native protein crops as a factor in improving food security in Poland. In Agrarian Perspectives XXVII.
Food Safety-Food Security, Proceedings of the 27th International Scientific Conference, 19-20
September 2018, Prague, Czech Republic (pp. 81-86). Czech University of Life Sciences Prague,
Faculty of Economics and Management.
[44] Ngo, V. M., Le-Khac, N. A., & Kechadi, M. (2018). An efficient data warehouse for crop yield
prediction. arXiv preprint arXiv:1807.00035.
[45] Aher, A., Kasar, J., Ahuja, P., &Jadhav, V. (2018). Smart agriculture using clustering and IOT.
International Research Journal of Engineering and Technology (IRJET), 5(03), 4065-4068.
[46] Coble, K. H., Mishra, A. K., Ferrell, S., & Griffin, T. (2018). Big data in agriculture: A challenge
for the future. Applied Economic Perspectives and Policy, 40(1), 79-96.
[47] Kovács, I., & Husti, I. (2018). The role of digitalization in the agricultural 4.0–how to connect the
industry 4.0 to agriculture?. Hungarian Agricultural Engineering, 33(1), 38-42.
[48] Kumar, H., & Menakadevi, T. (2018). A review on big data analytics in the field of
agriculture. International Journal of Latest Transactions in Engineering and Science, 1(4), 1-10.
[49] Santhiya, B., Govindharaj, I., Soundharya, C., Priyanka, R., Rajadurai, R., Maduravani, B., &
Shanmukhi, M. (2018). Smart Monitoring of Tarrace Garden using Solar Energy. International Journal
of Pure and Applied Mathematics, 118(14), 387-392.
[50] Sarker, M. N. I., Islam, M. S., Ali, M. A., Islam, M. S., Salam, M. A., & Mahmud, S. H. (2019).
Promoting digital agriculture through big data for sustainable farm management. International Journal
of Innovation and Applied Studies, 25(4), 1235-1240.
[51] Priya, P. L. V., Harshith, N. S., & Ramesh, D. N. (2018). Smart agriculture monitoring system
using IoT. International Journal of Engineering & Technology, 7(2.7), 308-311.
[52] Mathivanan, S., & Jayagopal, P. (2019). A Big Data Virtualization Role in Agriculture: A
Comprehensive Review. Walailak Journal of Science and Technology (WJST), 16(2), 55-70.
[53] Surya, P., & Laurence Aroquiaraj, I. (2019). Performance Analysis of K-Means and K-Medoid
Clustering Algorithms Using Agriculture Dataset. Journal of Emerging Technologies and Innovative
Research (JETIR), 6(1), 539-545.
[54] Anushree, M. K., & Krishna, R. (2018). A smart farming system using Arduino based
technology. Int. J. Adv. Res. Ideas Innov. Technol, 4(4), 850-856.
[55] Liu, S., Guo, L., Webb, H., Ya, X., & Chang, X. (2019). Internet of Things monitoring system of
modern eco-agriculture based on cloud computing. IEEE Access, 7(1), 37050-37058.
[56] Chen, J., & Yang, A. (2019). Intelligent agriculture and its key technologies based on internet of
things architecture. IEEE Access, 7(1), 77134-77141.
[57] Kamath, R., Balachandra, M., & Prabhu, S. (2019). Raspberry pi as visual sensor nodes in precision
agriculture: A study. IEEE Access, 7(1), 45110-45122.
[58] Hu, Z., Xu, L., Cao, L., Liu, S., Luo, Z., Wang, J., ...& Wang, L. (2019). Application of non-
orthogonal multiple access in wireless sensor networks for smart agriculture. IEEE Access, 7(1), 87582-
87592.
[59] Tseng, F. H., Cho, H. H., & Wu, H. T. (2019). Applying big data for intelligent agriculture-based
crop selection analysis. IEEE Access, 7(1), 116965-116974.
[60] Farooq, M. S., Riaz, S., Abid, A., Abid, K., &Naeem, M. A. (2019). A Survey on the Role of IoT
in Agriculture for the Implementation of Smart Farming. IEEE Access, 7(1), 156237-156271.
[61] Tóth, M., Felföldi, J., &Szilágyi, R. (2019). Possibilities of IoT based management system in
greenhouses. Georgikon for Agriculture, 23(3), 43-62.
[62] Rao, G. P. R., Indira, V. V. S., Manikanta, P., &Srinivas, D. M. (2019). Large Scale Farming
Analysis with the Help of IOT & Data Analytics. International Journal of Advanced Multidisciplinary
Scientific Research (IJAMSR), 2(3), 27-39.
[63] Rasooli, M. W., Bhushan, B., & Kumar, N. (2020). Applicability of wireless sensor networks &
IoT in saffron & wheat crops: A smart agriculture perspective. Int. J. Scientifi c Technol. Res, 9(2),
2456-2461.
[64] Pini, M., Marucco, G., Falco, G., Nicola, M., & De Wilde, W. (2020). Experimental Testbed and
Methodology for the Assessment of RTK GNSS Receivers Used in Precision Agriculture. IEEE
Access, 8(1), 14690-14703.
[65] Saad, A., & Gamatié, A. (2020). Water Management in Agriculture: A Survey on Current
Challenges and Technological Solutions. IEEE Access, 8(1), 38082-38097.
[66] Gupta, M., Abdelsalam, M., Khorsandroo, S., & Mittal, S. (2020). Security and privacy in smart
farming: Challenges and opportunities. IEEE Access, 8(1), 34564-34584.
[67] Liu, R., Zhang, Y., Ge, Y., Hu, W., & Sha, B. (2020). Precision Regulation Model of Water and
Fertilizer for Alfalfa Based on Agriculture Cyber-Physical System. IEEE Access, 8(1), 38501-38516.
[68] Gulec, O., Haytaoglu, E., & Tokat, S. (2020). A Novel Distributed CDS Algorithm for Extending
Lifetime of WSNs With Solar Energy Harvester Nodes for Smart Agriculture Applications. IEEE
Access, 8(12), 58859-58873.
[69] Huang, K., Shu, L., Li, K., Yang, F., Han, G., Wang, X., & Pearson, S. (2020). Photovoltaic
Agricultural Internet of Things Towards Realizing the Next Generation of Smart Farming. IEEE
Access, 8(22), 76300-76312.
[70] Balamurugan, S., Divyabharathi, N., Jayashruthi, K., Bowiya, M., Shermy, R. P., & Shanker, R.
(2016). Internet of agriculture: Applying IoT to improve food and farming technology. International
Research Journal of Engineering and Technology (IRJET), 3(10), 713-719.
[71] Himesh, S., Prakasa Rao, E. V. S., Gouda, K. C., Ramesh, K. V., Rakesh, V., & Mohapatra, G. N.
(2018). Digital revolution and Big Data: a new revolution in agriculture. CAB Rev, 13(21), 1-7.
[72] Corcoran, P. (2014). Smartphones and the Internet of Servants [Notes from the Editor]. IEEE
Consumer Electronics Magazine, 3(4), 4-9.
[73] Soman, S., Byju, G., & Bharathan, R. (2013). GIS based decision support system for precision
farming of cassava in India. ActaBiologicaIndica, 2(2), 394-399.
[74] Shekhar, S., Schnable, P., LeBauer, D., Baylis, K., & Waal, K. V. (2017). Agriculture Big Data
(AgBD) Challenges and Opportunities from Farm to Table: A Midwest Big Data Hub Community
Whitepaper. White paper for the US National Institute of Food and Agriculture, 1-12.
[75] Morota, G., Ventura, R. V., Silva, F. F., Koyama, M., & Fernando, S. C. (2018). Big data analytics
and precision animal agriculture symposium: machine learning and data mining advance predictive big
data analysis in precision animal agriculture. Journal of animal science, 96(4), 1540-1550.
[76] Astill, J., Dara, R. A., Fraser, E. D., & Sharif, S. (2018). Detecting and predicting emerging disease
in poultry with the implementation of new technologies and big data: a focus on Avian Influenza
Virus. Frontiers in veterinary science, 5(2), 1-12.
[77] Kheyfets, B. A., & Chernova, V. Y. (2019). Sustainable agriculture in Russia: Research on the
dynamics of innovation activity and labor productivity. Entrep. Sustain, 7(6), 814-824.
[78] Shekhar, S., Colletti, J., Muñoz-Arriola, F., Ramaswamy, L., Krintz, C., Varshney, L., &
Richardson, D. (2017). Intelligent infrastructure for smart agriculture: An integrated food, energy and
water system. 1(1), 1-8.
[79] Lee, S. W., Sarp, S., Jeon, D. J., & Kim, J. H. (2015). Smart water grid: the future water
management platform. Desalination and Water Treatment, 55(2), 339-346.
[80] Böröcz, M., Szőke, L., & Horváth, B. (2016). Possible climate friendly innovation ways and
technical solutions in the agricultural sector for 2030. Hungarian Agriculture System, 29(11), 55-59.
[81] Sanders, K. T., & Masri, S. F. (2016). The energy-water agriculture nexus: the past, present and
future of holistic resource management via remote sensing technologies. Journal of Cleaner
Production, 117(23), 73-88.
[82] Mekonnen, Y., Burton, L., Sarwat, A., & Bhansali, S. (2018, October). IoT sensor network
approach for smart farming: An application in food, energy and water system. In 2018 IEEE Global
Humanitarian Technology Conference (GHTC) 1-5. IEEE.
[83] Burton, L., Dave, N., Fernandez, R. E., Jayachandran, K., & Bhansali, S. (2018). Smart gardening
IoT soil sheets for real-time nutrient analysis. Journal of The Electrochemical Society, 165(8), B3157.
[84] Gayatri, M. K., Jayasakthi, J., & Mala, G. A. (2015, July). Providing Smart Agricultural solutions
to farmers for better yielding using IoT. In 2015 IEEE Technological Innovation in ICT for Agriculture
and Rural Development (TIAR) 40-43. IEEE.
[85] Gondchawar, N., & Kawitkar, R. S. (2016). IoT based smart agriculture. International Journal of
advanced research in Computer and Communication Engineering, 5(6), 838-842.
[86] hadoop.apache.org. (2020). Hadoop HDFS Architecture [online] Available at:
https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/images/hdfsarchitecture.png.
[Accessed 10 March. 2020].
[87] Shahzadi, R., Tausif, M., Ferzund, J., & Suryani, M. A. (2016). Internet of things based expert
system for smart agriculture. International Journal of Advanced Computer Science and
Applications, 7(9), 341-350.
[88] Jawad, H. M., Nordin, R., Gharghan, S. K., Jawad, A. M., & Ismail, M. (2017). Energy-efficient
wireless sensor networks for precision agriculture: A review. Sensors, 17(8), 1781-1790.
[89] Suma, N., Samson, S. R., Saranya, S., Shanmugapriya, G., & Subhashri, R. (2017). IOT based
smart agriculture monitoring system. International Journal on Recent and Innovation Trends in
computing and communication, 5(2), 177-181.
[90] a4academics.com. (2020). Hadoop-Mapreduce-Architecture [online] Available at:
http://a4academics.com/images/hadoop/Hadoop-Mapreduce-Architecture.png [Accessed 9 March.
2020].
[91] Aithal, P. S., (2016). Study on ABCD Analysis Technique for Business Models, Business
Strategies, Operating Concepts & Business Systems. International Journal in Management and Social
Science, 4(1), 98-115.
[92] Aithal, P. S., Shailashree, V., & Kumar, P. M. (2015). A new ABCD technique to analyze business
models & concepts. International Journal of Management, IT and Engineering, 5(4), 409-423.
**************